From kimsangb at stanford.edu Tue Feb 17 22:53:01 2009 From: kimsangb at stanford.edu (SangBum Kim) Date: Tue, 17 Feb 2009 22:53:01 -0800 (PST) Subject: overexposed pattern for 1um 3612 Message-ID: <198374663.717541234939981234.JavaMail.root@zm01.stanford.edu> Dear karlsuss2 users, It has been reported that 1um 3612 exposed for 1.0s with Vac contact mode in Karlsuss2 is 'overexposed', https://spf.stanford.edu/SNF/processes/process-modules/photolithography/photolithography-qualifications-1/photolithography-qualifications Has anyone been able to achieve appropriate exposure by decreasing the exposure time to be less than 1.0? >From what I observed from overexposed patterns on my wafer which was exposed for 1.1s with LVac contact mode, I was not sure whether the overexposure was caused either by a gap between the mask and the wafer or by the long exposure time. (My 3um by 3um squares look like almost circles.) Thanks, SangBum From mahnaz at stanford.edu Wed Feb 18 10:01:40 2009 From: mahnaz at stanford.edu (Mahnaz Mansourpour) Date: Wed, 18 Feb 2009 10:01:40 -0800 Subject: overexposed pattern for 1um 3612 In-Reply-To: <198374663.717541234939981234.JavaMail.root@zm01.stanford.edu> References: <198374663.717541234939981234.JavaMail.root@zm01.stanford.edu> Message-ID: <499C4D04.6010705@stanford.edu> Hello SangBum, You need to realize that the exposer will need to adjust with regards to your mask polarity and substrate. Not knowing any of this information, for 1 um resist on silicon with mask mostly chrome you will need somewhere from 1.1 to 1.3 seconds of exposer. This will change if you have metals on the wafer, you will need less exposer. On the other hand, with silicon and 1 um of resist but with mostly open mask ( light field) you will need less exposer say .8 to 1 seconds of exposer. Doing a test wafer is always required. although I have asked the intensity of the systems to be checked twice a week but labmembers are responsible to do a test wafers. From my end I will check to see if the intensity checks were done. I hope this help but please come and see me as I am confused about your email. mahnaz SangBum Kim wrote: > Dear karlsuss2 users, > > It has been reported that 1um 3612 exposed for 1.0s with Vac contact mode in Karlsuss2 is 'overexposed', > > https://spf.stanford.edu/SNF/processes/process-modules/photolithography/photolithography-qualifications-1/photolithography-qualifications > > Has anyone been able to achieve appropriate exposure by decreasing the exposure time to be less than 1.0? > > From what I observed from overexposed patterns on my wafer which was exposed for 1.1s with LVac contact mode, I was not sure whether the overexposure was caused either by a gap between the mask and the wafer or by the long exposure time. (My 3um by 3um squares look like almost circles.) > > Thanks, > SangBum > From kimsangb at stanford.edu Thu Feb 19 00:57:57 2009 From: kimsangb at stanford.edu (SangBum Kim) Date: Thu, 19 Feb 2009 00:57:57 -0800 (PST) Subject: overexposed pattern for 1um 3612 In-Reply-To: <1660777871.982001235033455742.JavaMail.root@zm01.stanford.edu> Message-ID: <1084284291.982431235033877253.JavaMail.root@zm01.stanford.edu> Thank you for those who responded to my email and Mahnaz for helping me during my exposure. I did some test run today and it turns out that the exposure result really depends on which wafer chuck I use. There are 3 wafer chucks and one of them gave me unacceptable overexposure results making squares look like almost circles. The other two chucks were relatively OK because they also resulted in overexposure in some areas of my wafer. I don't think that lamp intensity variation across the wafer is responsible for non-uniform exposure result since those overexposed areas on a wafer changes when I change the chuck. I am not sure whether those chucks will give me consistently bad results since I ran only one wafer from each chuck. However, if the chuck is really causing this, does it mean that we should replace or fix those chucks that do not work so well? Or is there any thing else that I can do to prevent this from happening again? FYI, my wafers were blanket wafers with 30nm thermal oxide, LOL2000, 1um 3612 resists on them. Thanks, SangBum ----- Original Message ----- From: "SangBum Kim" To: karlsuss2 at snf.stanford.edu Sent: Tuesday, February 17, 2009 10:53:01 PM GMT -08:00 US/Canada Pacific Subject: overexposed pattern for 1um 3612 Dear karlsuss2 users, It has been reported that 1um 3612 exposed for 1.0s with Vac contact mode in Karlsuss2 is 'overexposed', https://spf.stanford.edu/SNF/processes/process-modules/photolithography/photolithography-qualifications-1/photolithography-qualifications Has anyone been able to achieve appropriate exposure by decreasing the exposure time to be less than 1.0? >From what I observed from overexposed patterns on my wafer which was exposed for 1.1s with LVac contact mode, I was not sure whether the overexposure was caused either by a gap between the mask and the wafer or by the long exposure time. (My 3um by 3um squares look like almost circles.) Thanks, SangBum From jimkruger at yahoo.com Thu Feb 19 07:10:21 2009 From: jimkruger at yahoo.com (jim kruger) Date: Thu, 19 Feb 2009 07:10:21 -0800 (PST) Subject: overexposed pattern for 1um 3612 In-Reply-To: <1084284291.982431235033877253.JavaMail.root@zm01.stanford.edu> Message-ID: <523198.37324.qm@web38905.mail.mud.yahoo.com> Without looking at the wafers, I suspect that the problem here is not of exposure level but of varying mask contact. That is why it varies chuck to chuck. Ask staff to clean the chucks and make sure your wafers are flat and clean. Inspect each wafer after exposure for missing resist spots and re-clean the mask, perhaps even before each wafer contact. If you really need square corners on 3 um CDs, you might consider using the ASML. jim --- On Thu, 2/19/09, SangBum Kim wrote: > From: SangBum Kim > Subject: Re: overexposed pattern for 1um 3612 > To: karlsuss2 at snf.stanford.edu > Date: Thursday, February 19, 2009, 12:57 AM > Thank you for those who responded to my email and Mahnaz for > helping me during my exposure. > > I did some test run today and it turns out that the > exposure result really depends on which wafer chuck I use. > There are 3 wafer chucks and one of them gave me > unacceptable overexposure results making squares look like > almost circles. The other two chucks were relatively OK > because they also resulted in overexposure in some areas of > my wafer. I don't think that lamp intensity variation > across the wafer is responsible for non-uniform exposure > result since those overexposed areas on a wafer changes when > I change the chuck. > > I am not sure whether those chucks will give me > consistently bad results since I ran only one wafer from > each chuck. However, if the chuck is really causing this, > does it mean that we should replace or fix those chucks that > do not work so well? Or is there any thing else that I can > do to prevent this from happening again? > > FYI, my wafers were blanket wafers with 30nm thermal oxide, > LOL2000, 1um 3612 resists on them. > > Thanks, > SangBum > > > ----- Original Message ----- > From: "SangBum Kim" > To: karlsuss2 at snf.stanford.edu > Sent: Tuesday, February 17, 2009 10:53:01 PM GMT -08:00 > US/Canada Pacific > Subject: overexposed pattern for 1um 3612 > > Dear karlsuss2 users, > > It has been reported that 1um 3612 exposed for 1.0s with > Vac contact mode in Karlsuss2 is 'overexposed', > > https://spf.stanford.edu/SNF/processes/process-modules/photolithography/photolithography-qualifications-1/photolithography-qualifications > > Has anyone been able to achieve appropriate exposure by > decreasing the exposure time to be less than 1.0? > > From what I observed from overexposed patterns on my wafer > which was exposed for 1.1s with LVac contact mode, I was not > sure whether the overexposure was caused either by a gap > between the mask and the wafer or by the long exposure time. > (My 3um by 3um squares look like almost circles.) > > Thanks, > SangBum